You are on page 1of 14

Estuaries and Coasts (2019) 42:2170–2183

https://doi.org/10.1007/s12237-019-00626-x

Fish Diet Shifts Associated with the Northern Gulf of Mexico


Hypoxic Zone
Cassandra N. Glaspie 1 & Melissa Clouse 2 & Klaus Huebert 3 & Stuart A. Ludsin 4 & Doran M. Mason 5 & James J. Pierson 3 &
Michael R. Roman 3 & Stephen B. Brandt 6

Received: 26 September 2017 / Revised: 11 August 2019 / Accepted: 21 August 2019 / Published online: 28 October 2019
# Coastal and Estuarine Research Federation 2019

Abstract
The occurrence of low dissolved oxygen (hypoxia) in coastal waters may alter trophic interactions within the water column. This
study identified a threshold at which hypoxia in the northern Gulf of Mexico (NGOMEX) alters composition of fish catch and
diet composition (stomach contents) of fishes using fish trawl data from summers 2006–2008. Hypoxia in the NGOMEX
impacted fish catch per unit effort (CPUE) and diet below dissolved oxygen thresholds of 1.15 mg L−1 (for fish CPUE) and
1.71 mg L−1 (for diet). CPUE of many fish species was lower at hypoxic sites (≤ 1.15 mg L −1) as compared to normoxic regions
(> 1.15 mg L −1), including the key recreational or commercial fish species Atlantic croaker Micropogonias undulatus and red
snapper Lutjanus campechanus. Overall, fish diets from hypoxic sites (≤ 1.71 mg L−1) and normoxic sites (> 1.71 mg L−1)
differed. Fish caught in normoxic regions consumed a greater mass of benthic prey (ex. gastropods, polychaetes) than fish caught
in hypoxic regions. Hypoxia may increase predation risk of small zooplankton, with observations of increased mass of small
zooplankton in fish stomachs when bottom hypoxia was present. Changes in contributions of small zooplankton and benthic prey
to fish diet in hypoxic areas may alter energy flow in the NGOMEX pelagic food web and should be considered in fishery
management.

Keywords Fish diet . Dissolved oxygen . Predation . Fishery

Introduction identified worldwide, with an area exceeding 20,000 km2 in


some years (Turner et al. 2008; Bianchi et al. 2010). The
Seasonal bottom hypoxia has emerged as one of the major occurrence and size of the zone is a result of high nutrient
global problems in freshwater, estuarine, and coastal marine loading from the heavily agricultural Mississippi River water-
ecosystems (Diaz and Rosenberg 2008). One of the most shed (Turner et al. 2008; Bianchi et al. 2010). The effects of
widely known reoccurring summer hypoxic zones exists in hypoxia on NGOMEX living resources are of particular inter-
the northern Gulf of Mexico (NGOMEX). The NGOMEX est, not only because of the extent of hypoxia, but also given
hypoxic zone is one of the largest areas of coastal hypoxia the economic importance of this region for commercial and

Communicated by Wim J. Kimmerer


Electronic supplementary material The online version of this article
(https://doi.org/10.1007/s12237-019-00626-x) contains supplementary
material, which is available to authorized users.

* Cassandra N. Glaspie 3
Horn Point Laboratory, University of Maryland Center for
clgaspie1@lsu.edu Environmental Science, PO Box 775, Cambridge, MD 21613, USA
4
Aquatic Ecology Laboratory, Department of Evolution, Ecology, and
1
Department of Oceanography and Coastal Sciences, Louisiana State Organismal Biology, Ohio State University, 1314 Kinnear Road,
University, 3195 Energy, Coast, and Environment Building, Baton Columbus, OH 43212, USA
Rouge, LA 70803, USA 5
NOAA Great Lakes Environmental Research Laboratory, 4840 S.
State Road, Ann Arbor, MI 48108, USA
2 6
School of Pharmacy, Hampton University, Kittrell Hall, Department of Fisheries and Wildlife, Oregon State University, 104
Hampton, VA 23668, USA Nash Hall, Corvallis, OR 97331, USA
Estuaries and Coasts (2019) 42:2170–2183 2171

recreational fishing, and the efforts and costs to control the largely hypothetical (Costantini et al. 2008; Arend et al.
size of the hypoxic zone through landscape/watershed man- 2011; Brandt et al. 2011; Zhang et al. 2014). Many studies
agement and nutrient reduction (Rabotyagov et al. 2014). have documented diets of NGOMEX species (Sutton and
Effects of hypoxia on fish may occur through direct and Hopkins 1996; Bethea et al. 2007; Wells et al. 2008), but
indirect processes including changes in spatial distributions few estimate how fish diet may be altered in hypoxic areas
(Ludsin et al. 2009), reproduction and recruitment (Shang (though see Aku and Tonn 1999; Pothoven et al. 2009).
and Wu 2004; Thomas and Rahman 2012), vital rates (e.g., Increasingly, simulations and modeling studies that incorpo-
growth and mortality), and increased susceptibility to other rate several components of the marine food web have been
stressors (Breitburg et al. 2009). Examples of direct effects used to examine increases or decreases in fisheries production
include reduced fish catch per unit effort (CPUE) in the under various scenarios of hypoxia severity (de Mutsert et al.
Chesapeake Bay (Buchheister et al. 2013) and in the northeast 2016; Rose et al. 2018). However, these studies are limited by
Pacific (Hughes et al. 2015) and increased CPUE of menha- the available knowledge of fish diet, especially for some com-
den in the NGOMEX (Langseth et al. 2014). Reductions in mon forage species such as Atlantic bumper Chloroscombrus
abundance of sensitive fish species occur due to fish kills chrysurus (Glaspie et al. 2018).
(Thronson and Quigg 2008) or changes in the spatial distribu- Assessing thresholds of dissolved oxygen at which
tion, with mobile species avoiding low oxygen waters, but sublethal or lethal effects occur for a particular species
occurring above or congregating at the horizontal edges of or community of organisms is essential to manage marine
hypoxic regions (Craig and Crowder 2005; Hazen et al. systems experiencing hypoxia. This information can be
2009; Ludsin et al. 2009; Zhang et al. 2009). Concentration used to predict when fisheries will fail (Renaud 1986) or
of fish in small pockets of suitable habitat, or changes in fish to set targets to avoid mortality of fish and invertebrates
behavior due to hypoxia, may increase fish CPUE by render- (Vaquer-Sunyer and Duarte 2008). In the literature, hyp-
ing fish more susceptible to fishing gear (Breitburg et al. 2009; oxia thresholds typically refer to bottom dissolved oxygen
Langseth et al. 2014). levels ≤ 2 mg L−1 (Renaud 1986). However, in laboratory
Hypoxia-induced changes in food webs result from shifts studies, the median lethal oxygen concentration (LC50) for
in the abundance and spatial distribution of lower trophic major groups of marine organisms varies from 0.89
levels (Breitburg et al. 1997; Ekau et al. 2010; Roman et al. (gastropods) to 2.45 mg L −1 (crustaceans) (Vaquer-
2019). Hypoxia is associated with high zooplankton mortality Sunyer and Duarte 2008). This indicates that thresholds
and low zooplankton biomass (Kimmel et al. 2009). Sensitive other than 2 mg L−1 may be more meaningful for fish and
species include some commonly found in the NGOMEX such invertebrate communities. Few studies have examined
as the copepods Acartia tonsa (Elliott et al. 2013), hypoxia thresholds in situ (through see Eby and
Centropages hamatus (Stalder and Marcus 1997), Crowder 2002). Alternative hypoxia thresholds for fish
Paracalanus sp., and Oithona sp. (Zhang and Wong 2011). species in the NGOMEX have not been assessed, nor
Small zooplankton in particular may be more susceptible to have thresholds incorporating hypoxia-related changes in
low oxygen; in the Chesapeake Bay, hypoxia has been asso- diet composition. An improved ability to understand how
ciated with zooplankton communities composed of large in- hypoxia influences foraging interactions between fish and
dividuals (Kimmel et al. 2009). zooplankton in the NGOMEX should generally benefit
In addition to the direct effects of hypoxia on specific taxa, our ability to model and forecast the long-term conse-
hypoxia can alter trophic interactions by affecting predator or quences of hypoxia on pelagic fish populations and fish-
prey escape/capture responses. For example, hypoxia de- eries productivity, which has thus far remained elusive in
creases prey escape response and increases efficiency of cap- nearly all ecosystems (Rose et al. 2004; Breitburg et al.
ture by predators (Breitburg et al. 1997; Decker et al. 2004; 2009; Hazen et al. 2009). Here, we describe changes in
Domenici et al. 2007); hypoxia can also reduce (Keister et al. the diets of fish and the structure of the pelagic food web
2000; Taylor and Rand 2003; Ludsin et al. 2009) or increase relative to the occurrence of hypoxia in the NGOMEX.
(Prince and Goodyear 2006; Costantini et al. 2008) the spatial Water column dissolved oxygen, fish CPUE and spa-
overlap between predators and prey. Observations of hypoxia- tial distribution, and fish diet composition data collected
related changes in the spatial distribution of fish, the vertical in the NGOMEX during 2006–2008 were used to iden-
distribution of zooplankton, and the size structure of the zoo- tify thresholds of bottom DO below which fish CPUE
plankton community (Zhang et al. 2009; Kimmel et al. 2010; and diet (stomach contents) were altered. Hypoxia
Roman et al. 2012) all suggest that altered trophic interactions thresholds were then used to (1) examine the effects of
are an important ecological consequence of hypoxia for zoo- hypoxia on fish CPUE and (2) determine if diet compo-
plankton and their predators in the NGOMEX. sition differs for fish caught in normoxic and hypoxic
The relationship between hypoxia and trophic dynamics areas for zooplanktivorous, benthivorous, and piscivo-
(e.g., zooplanktivory, benthivory, and piscivory) remains rous fish.
2172 Estuaries and Coasts (2019) 42:2170–2183

Methods To determine fish diet composition, a minimum of 15 non-


empty fish stomachs per species were analyzed from each
Sample Collection trawl station whenever possible. Fish were thawed and total
length (TL) measured to the nearest 1 mm. Stomachs were
Samples were collected from the NGOMEX aboard the R/V removed and dissected under a microscope. Fish were dried
Pelican (Louisiana Universities Marine Consortium, in a drying oven and weighed to the nearest 0.0001 g dry
LUMCOM), with cruises on August 4–13, 2006, July 30– mass. All zooplankton in stomachs were identified to the low-
October 14, 2007, and August 1–August 11, 2008 (Fig. 1). est possible taxon and counted using a dissecting microscope.
Physical properties of the water column, including tempera- A minimum of 50 individuals in each taxon were digitized and
ture and dissolved oxygen, were measured with a CTD measured to the nearest 0.01 mm with ImagePro Plus (Media
(Seabird SBE 9 with a SBE 43 dissolved oxygen probe). Cybernetics, Inc. Silver Spring, MD). Partial animals were
Fish were collected using a bottom trawl (7.62 m head- counted as individuals, but not measured for length. Lengths
rope, 3.66 m mouth depth; 38 mm stretch mesh; 12 mm of zooplankton were converted to dry mass using relationships
cod-end liner) or a mid-water trawl (9.14 m wide, 6.10 m tall, reported in the literature (Fontaine and Neal 1971; Uye 1982;
12 mm cod-end liner). Trawling occurred day and night, and Cadman and Weinstein 1985; Chisholm and Roff 1990;
trawl duration varied between 10 and 60 min to ensure ade- Webber and Roff 1995; Hopcroft et al. 1998; Tita et al.
quate collection of fish. After capture, fish were identified, 1999; Ara 2001; Remsen et al. 2004; Rose et al. 2004).
counted, and frozen at − 20 °C. Because the bottom trawl Mean dry mass of individuals in each category was multiplied
net was not opening and closing, resulting samples may have by the total number to calculate the total dry mass for each
included fish from higher in the water column that would have prey category. Dry mass of stomach contents was divided by
been captured during net deployment and retrieval. Trawl fish total dry mass (g) to account for differences in fish size,
times used to calculate an index of fish abundance, catch per and thus stomach capacity.
unit effort (CPUE, number of fish min−1), reflected only the Due to the wide range of prey species examined, fish spe-
amount of time the bottom trawl was on the shelf bottom or at cies were classified as zooplanktivores, benthivores, or
mid-water targeted depth. piscivores for analysis. We used k-means clustering with the

Fig. 1 Trawling sites in the Northern Gulf of Mexico, 2006–2008. Symbols denote years, shading denotes bottom dissolved oxygen availability, with
hypoxic areas ≤ 1.71 mg L−1 in black and normoxic areas > 1.71 mg L−1 in gray
Estuaries and Coasts (2019) 42:2170–2183 2173

Hartigan–Wong algorithm (Hartigan and Wong 1979) to par- trawl. Only fish taxa or stomach content taxa that appeared in
tition the species into three groups based on an 8 column ≥ 10 trawls were used to complete this analysis.
matrix summarizing their stomach contents. Each prey item
in stomach contents was categorized as either small zooplank- Diet Composition Analysis
ton (≤ 5 μg mean dry mass), large zooplankton (> 5 μg mean
dry mass), benthic organisms, or mobile prey (fish, shrimp, To test if diet composition differed between hypoxic and
and squid). The first four columns of the stomach content normoxic areas, stomach content composition data were ana-
matrix were the proportion of prey in each category (mean lyzed using PERMANOVA (Anderson 2008). The
mass from each category divided by mean total mass). The community-level change point xcp (threshold) for fish stom-
next four columns were the frequency of occurrence of each ach contents was used to assign the category “hypoxic” or
category. The bootstrapped Jaccard coefficient was calculated “normoxic” to each trawl, depending on the bottom DO at
to assess fit of the k-means groupings (Hennig 2007). K- each trawl location, as determined from CTD data.
means clustering was only completed on taxa for which at PERMANOVA analysis included only taxa for which at least
least 10 fish were processed in both normoxic and hypoxic 10 fish were processed in both normoxic and hypoxic areas
areas, which included 14 fish species (Table 1). Clustering (Table 1). The PERMANOVA was completed using Bray
identified one grouping of six species that consumed small Curtis dissimilarity matrices calculated from fourth-root trans-
and large zooplankton more frequently than the other fish formed biomass (Anderson 2014), and models had the follow-
species analyzed, and this group was considered ing factors: dissolved oxygen (two levels: normoxic and hyp-
zooplanktivorous (Jaccard coefficient 0.89); one grouping of oxic), time of day the sample was collected (two levels: day
five species that frequently consumed benthic prey, but did not and night), diet classification (three levels: zooplanktivore,
often consume zooplankton or fish, and this group was con- benthivore, and piscivore), species (14 levels), and year (3
sidered benthivorous (Jaccard coefficient 0.81); and one levels). Bottom temperature was included as a covariable. If
grouping of three species that frequently consumed fish or multiple individuals of the same species were captured in a
highly mobile prey which was considered piscivorous single trawl, the mean prey biomass for all fish of that species
(Jaccard coefficient 0.77) (Table 1). These clusters were con- in the trawl was used to avoid pseudo-replication.
sidered stable (Jaccard coefficient 0.75–0.84) or highly stable Effects of spatial variability on the PERMANOVA results
(0.85–1.00) (Hennig 2008). were examined by running two additional PERMANOVA
models: the first included the sampling site latitude and lon-
gitude (normalized using z-score transformation; Anderson
Threshold Analysis 2005) as covariables, and the second included latitude/
longitude and sampling day nested within year (assuming
Hypoxic sampling stations were identified using Threshold trawls taken on the same day were more closely related, both
Indicator Taxa Analysis (TITAN) of stomach contents and in space and time). Repeating the analysis with covariables
bottom DO (Baker and King 2010). TITAN identified thresh- generated very similar results to those obtained from the orig-
olds in community data by combining change-point analysis inal PERMANOVA model, and only results from the original
(nCPA) with indicator species analysis. For each taxon in a model are shown here (Benedetti-Cecchi and Osio 2007).
community, the analysis produced a score (IndVal) estimating For all two-group comparisons, 95% confidence intervals
the association of the taxon to two groups separated at candi- were calculated using non-parametric bootstrap hypothesis
date change points (xi) along a gradient of a univariate indi- testing with 10,000 simulations (DiCiccio and Efron 1996).
cator variable, x (bottom DO). The IndVal for each taxon was All analyses were completed in R (R Core Team 2019). All
standardized as a z-score and the sum of z-scores for all taxa, data and code for this study have been archived and can be
sum(z), was calculated. The value of x that maximized sum(z) found at https://doi.org/10.5063/F10Z71M7.
was identified as a community-level change point, xcp.
Bootstrapped 95% confidence intervals for xcp are calculated
by resampling the observations (sampled with replacement, to Results
create a bootstrap sample the same size as the original dataset)
500 times. Threshold Analysis
We conducted TITAN analysis to identify thresholds
(change points) in (1) fish CPUE and (2) fish stomach con- A threshold in fish community composition was identified at
tents. CPUE was calculated for each fish species caught in bottom DO 1.15 mg L−1 (95% CI [0.99, 3.24]). A threshold in
each trawl. Fish stomach contents were calculated as the mean diet composition was identified at bottom DO 1.71 mg L−1
mass of each taxon found in stomach contents, divided by fish (95% CI [0.98, 3.33]). A threshold of 1.15 mg L−1 was used to
total dry mass (g), and averaged across all fish caught in each categorize bottom DO as “hypoxic” or “normoxic” for
2174 Estuaries and Coasts (2019) 42:2170–2183

Table 1 Total number of fish caught (No.) for zooplanktivorous (Z), benthivores (B), and piscivores (P) fish in the northern Gulf of Mexico

Species (common name) Diet No. PERM.

Anchoa hepsetus (striped anchovy) Z 549 X


Chloroscombrus chrysurus (Atlantic bumper) Z 610 X
Harengula jaguana (scaled sardine) Z 86 X
Leiostomus xanthurus (spot) Z 151 X
Lutjanus campechanus (red snapper) Z 144 X
Peprilus burti (gulf butterfish) Z 172 X
Larimus fasciatus (banded drum) B 78 X
Micropogonias undulatus (Atlantic croaker) B 920 X
Prionotus rubio (blackwing sea robin) B 172 X
Stenotomus caprinus (longspine porgy) B 200 X
Symphurus plagiusa (blackcheek tonguefish) B 71 X
Cynoscion arenarius (sand sea trout) P 415 X
Selene setapinnis (moonfish) P 65 X
Trichiurus lepturus (Atlantic cutlassfish) P 315 X
Anchoa mitchelli (bay anchovy) 40
Ariopsus felis (hardhead catfish) 30
Balistes capriscus (gray triggerfish) 54
Bregmaceros atlanticus (codlet) 13
Caranx crysos (blue runner) 36
Caranx hippos (crevalle jack) 2
Carcharhinus obscurus (dusky shark) 3
Centropristis philadelphica (rock sea bass) 7
Chaetodipterus faber (Atlantic spadefish) 1
Citharichthys spilopterus (bay whiff) 14
Decapterus punctatus (round scad) 17
Diplectrum bivittatum (dwarf sand perch) 28
Diplectrum formosum (regular sand perch) 1
Dorosoma petenense (threadfin shad) 5
Etropus crossotus (fringed flounder) 22
Gerreidae (mojarra) 6
Gobiidae (goby) 4
Gymnothorax nigromarginatus (black-edged moray eel) 8
Haemulon aurolineatum (tomtate) 2
Halieutichthys aculeatus (pancake batfish) 3
Kyphosus sectatrix (Bermuda chub) 2
Lagocephalus laevigatus (smooth puffer) 3
Lagodon rhomboides (pinfish) 17
Lepophidium brevibarbe (black-edged cusk eel) 24
Lutjanis synagris (lane snapper) 48
Menticirrhus americanus (kingfish) 1
Monacanthidae (filefish) 4
Ophichthidae (snake eel) 3
Ophidion welshi (crested cusk eel) 7
Opisthonema oglinum (Atlantic thread herring) 74
Peprilus paru (harvestfish) 2
Polydactylus octonemus (Atlantic threadfin) 3
Porichthys plectodon (Atlantic midshipman) 16
Rachycentron canadum (cobia) 7
Remora remora (remora) 2
Estuaries and Coasts (2019) 42:2170–2183 2175

Table 1 (continued)

Species (common name) Diet No. PERM.

Rhynchoconger flava (yellow conger eel) 12


Sardinella aurita (Spanish sardine) 58
Sciaenops ocellatus (red drum) 1
Scomber japonicus (chub mackerel) 2
Scomberomorus cavalla (king mackerel) 5
Seanus atrobranchus (blackear bass) 39
Seriola fasciata (lesser amberjack) 9
Seriola rivoliana (almaco jack) 15
Sphoeroides parvus (least puffer) 78
Sphyraena barracuda (barracuda) 4
Stellifer lanceolatus (star drum) 1
Syacium papillosum (dusky flounder) 55
Synodus poeyi (offshore lizardfish) 87
Upeneus parvus (dwarf goatfish) 5

PERMANOVA diet analysis (PERM.) is presented for the species with at least 10 stomachs processed for both normoxic and hypoxic areas, and these
species are indicated by an “X” in the last column

analysis of fish CPUE, and a threshold of 1.71 mg L−1 was sardine Sardinella aurita, and least puffer Sphoeroides parvus
used to categorize bottom DO as “hypoxic” or “normoxic” for (Fig. 2). The CPUE of all species listed in Table 1 can be
analysis of fish diet. found in Supplementary Fig. 1.
Hypoxia (≤ 1.71 mg L −1) was extensive throughout the
study period; across all 3 years, 29% of sites were hypoxic, Diet Composition Analysis
and bottom water dissolved oxygen ranged from 0.0 to
5.7 mg L−1 (Fig. 1). Mean bottom dissolved oxygen was Zooplankton in stomach contents included large zooplankton
2.6 mg L−1 (S.D. 1.2) in 2006, 2.8 mg L−1 (S.D. 0.9) in > 5 μg mean dry mass, such as Acartia sp., Centropages sp.;
2007, and 1.8 mg L−1 (S.D. 1.5) in 2008. Bottom temperature Eucalanus sp.; Temora sp.; cladocerans such as Evadne sp.,
ranged from 20.6 to 31.4 °C. Penilia sp., and Podon sp.; other calanoids such as
Clausocalanus sp., Labidocera sp., Pseudodiaptomus sp.,
Fish CPUE Undinula sp., Euchaeta sp., and Pontella sp.; barnacle larvae;
crab larvae; fish larvae; shrimp larvae; and urochordates.
The final dataset consisted of fish collected from n = 91 Small zooplankton < 5 μg mean dry mass in fish diets includ-
trawls over 1707 min in regions identified as normoxic ed Corycaeus sp., Oithona sp., Oncaea sp., Paracalanus sp.,
(> 1.15 mg L −1), and fish collected from n = 46 trawls over Saphirella sp., copepod nauplii, and harpacticoid copepods.
943 min in regions identified as hypoxic (≤ 1.15 mg L−1) Benthic organisms found in fish diets included amphipods,
(Fig. 1; Table 1). We found differences in fish species compo- bivalves, crabs, cumaceans, echinoderms, gastropods, iso-
sition and catch statistics between normoxic and hypoxic areas pods, mantis shrimp, nematodes, oligochaetes, ostracods,
of the NGOMEX (Fig. 2). The most abundant fish (in terms of polychaetes, and tanaids. Large, mobile prey, such as fish
CPUE) in both normoxic and hypoxic regions were striped and squid, was also found in fish diets.
anchovy Anchoa hepsetus, Atlantic bumper C. chrysurus, Large zooplankton made up a major portion of the diet for
sand seatrout Cynoscion arenarius, Atlantic croaker most species (Fig. 3). The most commonly found large zoo-
Micropogonias undulatus, and Atlantic cutlassfish plankton species in fish diets were shrimp larvae (found in
Trichiurus leptutus (Fig. 2). The CPUE of many species was 24% of fish stomachs, mean 1001 μg g−1 fish dry weight),
lower in hypoxic than in normoxic regions, including red Temora sp. (12% of stomachs, mean 5 μg g−1 fish dry weight),
snapper Lutjanus campechanus, Gulf butterfish Peprilus other calanoids (10% of stomachs, mean 20 μg g−1 fish dry
burti, M. undulatus, longspine porgy Stenotomus caprinus, weight), and Centropages sp. (10% of stomachs, mean
bay anchovy Anchoa mitchelli, gray triggerfish Balistes 6 μg g−1 fish dry weight). Other common prey items were
capriscus, dwarf sand perch Diplectrum bivittatum, pinfish benthic organisms (Fig. 3). The most commonly found ben-
Lagodon rhomboides, lane snapper Lutjanus synagris, thic species were nematodes (21% of fish stomachs, mean
Atlantic thread herring Opisthonema oglinum, Spanish 24 μg g−1 fish dry weight), polychaetes (15% of stomachs,
2176 Estuaries and Coasts (2019) 42:2170–2183

Fig. 2 Mean catch per unit effort (CPUE; number of fish min−1 trawl) of fish species in hypoxic (closed circles) and normoxic (open triangles) regions in
the northern Gulf of Mexico. Error bars are bootstrapped 95% confidence intervals

mean 203 μg g−1 fish dry weight), and gastropods (14% of urochordates, amphipods, mantis shrimp, and ostracods (Fig.
stomachs, mean 31 μg g−1 fish dry weight). Large, mobile 4). Several prey taxa had greater mass in the stomachs of
prey made up a substantial component of the diet for a few benthivores caught in normoxic areas, as compared to hypoxic
species, including C. arenarius, L. campechanus, and areas, including Paracalanus sp., Eucalanus sp., amphipods,
T. lepturus (Fig. 3). The most commonly found large, mobile gastropods, and polychaetes (Fig. 4). There were few oxygen-
prey groups were fish (7% of fish stomachs, mean 872 μg g−1 related differences in the mass of prey found in stomachs of
fish dry weight) and squid (2% of stomachs, mean 263 μg g−1 piscivores, though a greater mass of polychaetes was found in
fish dry weight). Small zooplankton made up a smaller com- the stomachs of piscivorous fish from normoxic regions, as
ponent of the diets of most fish species, although several zoo- compared to hypoxic regions (Fig. 4). Compared to fish diets
plankton taxa were commonly found in fish diets, including in hypoxic areas, there was a tendency for fish from all diet
Corycaeus sp. (17% of stomachs, mean 12 μg g−1 fish dry classes to consume greater mass of squid in normoxic areas,
weight), Paracalanus sp. (12% of stomachs, mean 11 μg g−1 and for benthivores to consume greater mass of fish in
fish dry weight), harpacticoid copepods (12% of stomachs, normoxic areas (Fig. 4).
mean 2 μg g−1 fish dry weight), and Oncaea sp. (10% of
stomachs, mean 5 μg g−1 fish dry weight). Means reported
are for all fish, not just those that had prey in stomachs. The
Discussion
diet of all species listed in Table 1 can be found in
Supplementary Fig. 2.
Community Thresholds
An interaction between dissolved oxygen and diet class
was identified with PERMANOVA (Table 2). A post hoc test
This study examined changes in fish diet composition and fish
was completed to interpret the main effect of dissolved oxy-
community composition relative to the occurrence of hypoxia
gen separately for each diet class. We conducted separate
in the NGOMEX. Hypoxia in the NGOMEX was associated
PERMANOVA analyses for zooplanktivores, benthivores,
with changes in fish catch per unit effort (CPUE) and diet
and piscivores. Dissolved oxygen was a significant variable
below dissolved oxygen thresholds of 1.15 mg L−1 (for fish
for zooplanktivores (F1,138 = 8.75, p = 0.001; Supplementary
CPUE) and 1.71 mg L−1 (for fish diet). The dissolved oxygen
Table 1), and benthivores (F 1 , 9 0 = 2.36, p = 0.03;
threshold for fish diet composition was higher than the thresh-
Supplementary Table 2), but not for piscivores (F1,71 = 1.53,
old for fish catch. As dissolved oxygen levels decline in the
p = 0.20; Supplementary Table 3). For zooplanktivores, great-
NGOMEX (especially off the coast of Texas, Karnauskas
er mass of some small zooplankton taxa (Oithona sp.,
et al. 2017), changes in the fish diet can be expected to occur
Paracalanus sp.) was found in the stomachs of fish caught
before changes in the fish community occur. Thus, changes in
in hypoxic areas, as compared to normoxic areas (Fig. 4). The
trophic transfer are a likely consequence of NGOMEX
mass of many other prey taxa was greater in the stomachs of
hypoxia.
zooplanktivorous fish caught in normoxic areas as compared
The thresholds for fish CPUE and fish diet in the
to those caught in hypoxic areas, including crab larvae,
NGOMEX were below the traditional threshold used to
Estuaries and Coasts (2019) 42:2170–2183 2177
2178 Estuaries and Coasts (2019) 42:2170–2183

ƒFig. 3 Diet composition for six zooplanktivorous fish species (a–f), five indicate that local fish catch is reduced when hypoxia is pres-
benthivorous fish species (g–k), and three piscivorous fish species (l–n) ent in the NGOMEX.
from samples taken in the northern Gulf of Mexico during summer of
2006–2008. Numbers above bars represent the total fish stomachs
Hypoxia-related decreases in catch may have implications
processed for that species, in either normoxic (Norm.) or hypoxic for fisheries management and conservation. Key commercial
(Hyp.) regions. Full species names can be found in Table 1 or recreational species were caught less often in hypoxic areas,
including Atlantic croaker and red snapper. Over 1 million
identify hypoxia (2 mg L−1). Studies using a threshold of Atlantic croaker are harvested annually in the recreational
2 mg L−1 may be missing the potential impacts of hypoxia fishery (NMFS 2017a). The red snapper fishery was worth
on the fish community and food web in this system. Many nearly $28 million in 2017 (NMFS 2018). Some of the species
species may forage near their metabolic limits at the hypoxic relatively absent from hypoxic areas, contributing to the lower
boundary, creating a hypoxic “edge effect” (Zhang et al. catch in those areas, are also species of concern. For example,
2009). It is important to consider the possibility that both red snapper and gray triggerfish have been identified as
community-level hypoxia thresholds in the NGOMEX may overfished species (NMFS 2017b). Gray triggerfish were rare-
be lower than in other systems. ly caught in trawls in hypoxic areas. Several species that had
lower CPUE in hypoxic regions, including Gulf butterfish,
bay anchovy, Atlantic thread herring, and Spanish sardine,
Fish CPUE are also key prey species for large predators in the
NGOMEX (Manooch and Hogarth 1983; Meyer and Franks
The relationship between hypoxia and pelagic fish catch is not 1996; Hoffmayer and Parsons 2003). Given the importance of
well-understood, and few studies have related fish catch to species such as red snapper to the economy of the NGOMEX,
hypoxia in other systems (Buchheister et al. 2013; Hughes future research on the relationship between hypoxia, local
et al. 2015). To our knowledge, this study is the first to relate displacements of fish, and population-level fishery trends or
catch of a suite of both demersal and pelagic fish species to catches is needed.
dissolved oxygen in the NGOMEX. The second objective of CPUE is often used as an index of abundance for fish
this study was to examine the effects of hypoxia on fish CPUE populations, but CPUE is not necessarily proportional to local
in the NGOMEX. In agreement with published literature abundance. CPUE depends upon catchability, which can in-
(Buchheister et al. 2013; Hughes et al. 2015), our results crease when fish are aggregated along the edge of the hypoxic
zone (Breitburg et al. 2009; Craig 2012; Langseth et al. 2014),
Table 2 PERMANOVA table indicating the effect of oxygen (two increasing CPUE without an increase in local abundance.
levels: normoxic and hypoxic), time of day the sample was collected Finally, even if CPUE and local abundance are proportional,
(two levels: day and night), diet classification (three levels:
zooplanktivore, benthivore, and piscivore), and species (14 levels) on we are unable to determine a mechanism for any changes in
diet composition of fish caught in the northern Gulf of Mexico fish abundance. Possible mechanisms for any decline in
CPUE may include mortality (direct or due to predation;
Df SumSq MeanSq F value P value
Thronson and Quigg 2008), vertical or horizontal migration
Time 1 0.14 0.14 0.9 0.52 (Craig and Crowder 2005; Hazen et al. 2009; Ludsin et al.
Diet class 2 7.31 3.66 24.17 0.001 2009; Zhang et al. 2009), or changes in reproduction or re-
Species 11 9.68 0.88 5.82 0.001 cruitment (Shang and Wu 2004; Thomas and Rahman 2012).
Oxygen 1 1.16 1.16 7.65 0.001
Bottom temperature 1 0.57 0.57 3.76 0.002
Year 2 1.24 0.62 4.1 0.001
Diet Composition
Time × diet class 2 0.47 0.24 1.56 0.07
Time × species 11 1.82 0.17 1.09 0.27
The third objective of this study was to determine if diet com-
position differs for fish caught in normoxic and hypoxic areas
Time × oxygen 1 0.07 0.07 0.45 0.89
for zooplanktivores, benthivores, and piscivores. There was a
Diet class × oxygen 2 0.82 0.41 2.71 0.001
significant impact of hypoxia on fish diet composition for both
Species × oxygen 11 1.83 0.17 1.1 0.25
zooplanktivores and benthivores. Fish caught in hypoxic areas
Time × diet class × oxygen 2 0.24 0.12 0.78 0.67
consumed less mass of large, mobile prey such as fish (for
Time × species × oxygen 10 1.15 0.11 0.76 0.94
benthivores) and squid (a tendency for all diet classes). This
Residuals 305 46.14 0.15
result can be explained by a distribution shift in mobile prey
Total 362 72.63
when hypoxia is present. Squid in particular are known to be
Significant p values (at α = 0.05) are italicized sensitive to hypoxia (Zielinski et al. 2000) and likely avoid
Df degrees of freedom, SumSq sum squared error, MeanSq mean squared hypoxic conditions. Small forage fishes such as juvenile an-
error chovies also avoid hypoxic waters (Taylor et al. 2007).
Estuaries and Coasts (2019) 42:2170–2183 2179

Fig. 4 Fourth-root transformed mean mass of prey in stomachs of fish processed in both normoxic and hypoxic areas). If multiple individuals of
from hypoxic (closed circles) and normoxic (open triangles) regions in the same species were captured in a single trawl, the mean prey biomass
the northern Gulf of Mexico. Means are calculated for zooplanktivores for all fish of that species in the trawl was used and the means presented
(top), benthivores (middle), and piscivores (bottom) for all fish species were not weighted by the number of fish caught in each trawl
included in PERMANOVA analysis (taxa for which at least 10 fish were
2180 Estuaries and Coasts (2019) 42:2170–2183

Hypoxia may result in increased predation risk for small et al. 2008). In contrast, lower CPUE of many small forage
zooplankton, since zooplanktivores in hypoxic areas con- fish, along with decreased consumption of fish, squid, and
sumed greater mass of small zooplankton than those in benthic organisms, may prevent nutrients from making their
normoxic areas. In the Chesapeake Bay, there was lower bio- way to higher consumers. This may alter food web structure in
mass of small zooplankton in hypoxic bottom water than in the NGOMEX, for example through increased dominance by
normoxic surface water, indicating a possible vertical distri- gelatinous zooplankton (Breitburg et al. 2003). The conse-
bution shift, and most (> 60%) of the zooplankton that were in quences of hypoxia and altered food web interactions remain
hypoxic waters were dead (Kimmel et al. 2009). Zooplankton speculative and have only been explored in modeling studies
moving out of the hypoxic waters may aggregate on the edges and small-scale laboratory studies (Breitburg et al. 1997;
of the hypoxic zone (Craig and Crowder 2005; Hazen et al. Brandt and Mason 2003; Breitburg et al. 2003; Brandt et al.
2009; Zhang et al. 2009), bringing dense concentrations of 2011; Zhang et al. 2014; de Mutsert et al. 2016); thus, further
zooplankton in contact with their predators. Several studies investigation into how localized shifts in distribution, growth,
have suggested that hypoxia (and subsequent habitat compres- and diets affect short-term or long-term growth in fish is
sion) of zooplankton may lead to increased predation by fish warranted.
(Vanderploeg et al. 2009a; Vanderploeg et al. 2009b; H. Management of NGOMEX fisheries in an ecosystem con-
Zhang et al. 2009; Brandt et al. 2011; Roman et al. 2012). A text requires ecosystem models that incorporate the impacts of
zooplankton distribution shift likely brings predators in closer hypoxia on interactions between fish and their prey. This
contact with zooplankton prey, and this effect may increase study is an important step to understanding local changes in
habitat quality for fish foraging in the surface waters of the fish food web dynamics in areas experiencing hypoxia, but
hypoxic zone. more research is needed to scale up to the entire NGOMEX,
Fish in hypoxic areas consumed less benthic prey such as and ultimately forecast the long-term effects of hypoxia on
gastropods and polychaetes. Previous studies in the pelagic fish populations. Such forecasts will be valuable to
Chesapeake Bay suggest that when hypoxia is intermittent understand changes in fisheries productivity, and to prevent
or moderate, some fish species consume more benthic prey catch limits from slipping into the realm of overfishing if
because benthic organisms such as clams and polychaetes hypoxia reduces fish survival, recruitment, or growth.
have been shown to reduce burial depth under hypoxia, mak- Incorporating hypoxia into ecosystem models and ultimately
ing them more susceptible to predation (Pihl et al. 1992; Long estimates of catch will improve fisheries management and
et al. 2008; Long and Seitz 2008). The low mass of benthic ensure these resources are available for future generations.
prey consumed by fish in hypoxic areas of the NGOMEX may
be a result of fish avoiding hypoxic bottom waters, as fish are Acknowledgements We thank James Roberts, Craig Stow, Stephen
rarely seen at oxygen concentrations below 2 mg L−1 Lozano, Jennifer Metes, Aly Peacy, and Katharine Bush for help with
collecting the field data. We also thank the captains and crew of the RV
(Rabalais and Turner 2001). However, fish are known to con- Pelican for their help with field sampling.
duct foraging forays into hypoxic bottom waters in many sys-
tems (Pihl et al. 1992; Rahel and Nutzman 1994; Roberts et al. Funding information This research was supported by NOAA-CSCOR
2009). Future research should focus on the availability of ben- Award NA06NOS4780148 and NA09NOS4780198 and National
Academies award NAS-GRP-2000006418. This is NOAA GLERL con-
thic prey resources in the NGOMEX hypoxic zone, and the
tribution No. 1931 and UMCES contribution No. 5713.
likelihood that some of the common fish species are able to
forage on benthic organisms for short periods under hypoxic
conditions.
Understanding how hypoxia might affect NGOMEX food References
web dynamics can help inform ecosystem models and help
Aku, Peter M.K., and William M. Tonn. 1999. Effects of hypolimnetic
agencies understand and predict how this ecologically and oxygenation on the food resources and feeding ecology of cisco in
economically important region might change with various Amisk Lake, Alberta. Transactions of the American Fisheries
hypoxia management scenarios. The results of this study sug- Society 128 (1): 17–30.
gest hypoxia may alter food web dynamics and trophic trans- Anderson, M.J. 2005. PERMANOVA: a FORTRAN computer program
for permutational multivariate analysis of variance. https://doi.org/
fer in the NGOMEX by decreasing fish CPUE and modifying 10.1002/9781118445112.stat07841.
fish diets. Increased consumption of zooplankton by fish may Anderson, Marti J. 2008. A new method for non-parametric multivariate
increase flow of energy to upper trophic levels in hypoxic analysis of variance. Austral Ecology 26 (1): 32–46.
regions, fueling increased growth and reproduction. Habitat Anderson, M. J. 2014. Permutational multivariate analysis of variance
compression of prey has been credited for increasing body (PERMANOVA). Wiley StatsRef: Statistics Reference Online.
https://doi.org/10.1002/9781118445112.stat07841, .
size of marlin and sailfish (Prince and Goodyear 2006) and Ara, Koichi. 2001. Length-weight relationships and chemical content of
providing an opportunity for population growth in the planktonic copepods in the Cananeia Lagoon estuarine system,
Chesapeake Bay striped bass Morone saxatilis (Costantini Sao Paulo, Brazil. Plankton Biology and Ecology 48: 121–127.
Estuaries and Coasts (2019) 42:2170–2183 2181

Arend, Kristin K., Dmitry Beletsky, Joseph DePinto, Stuart A. Ludsin, Craig, J. Kevin. 2012. Aggregation on the edge: Effects of hypoxia avoid-
James J. Roberts, Daniel K. Rucinski, Donald Scavia, David J. ance on the spatial distribution of brown shrimp and demersal fishes
Schwab, and Thomas O. Höök. 2011. Seasonal and interannual in the Northern Gulf of Mexico. Marine Ecology Progress Series
effects of hypoxia on fish habitat quality in Central Lake Erie. 445: 75–95. https://doi.org/10.3354/meps09437.
Freshwater Biology 56 (2): 366–383. https://doi.org/10.1111/j. Craig, J. Kevin, and Larry B. Crowder. 2005. Hypoxia-induced habitat
1365-2427.2010.02504.x. shifts and energetic consequences in Atlantic croaker and brown
Baker, Matthew E., and Ryan S. King. 2010. A new method for detecting shrimp on the Gulf of Mexico shelf. Marine Ecology Progress
and interpreting biodiversity and ecological community thresholds. Series 294: 79–94. https://doi.org/10.3354/meps294079.
Methods in Ecology and Evolution 1 (1): 25–37. https://doi.org/10. de Mutsert, Kim, Jeroen Steenbeek, Kristy Lewis, Joe Buszowski, James
1111/j.2041-210X.2009.00007.x. H. Cowan, and Villy Christensen. 2016. Exploring effects of hyp-
Benedetti-Cecchi, Lisandro, and Giacomo Chato Osio. 2007. Replication oxia on fish and fisheries in the northern Gulf of Mexico using a
and mitigation of effects of confounding variables in environmental dynamic spatially explicit ecosystem model. Ecological Modelling
impact assessment: Effect of marinas on rocky-shore assemblages. 331: 142–150. https://doi.org/10.1016/j.ecolmodel.2015.10.013.
Marine Ecology Progress Series 334: 21–35. https://doi.org/10. Decker, Mary Beth, Denise L. Breitburg, and Jennifer E. Purcell. 2004.
3354/meps334021. Effects of low dissolved oxygen on zooplankton predation by the
Bethea, Dana M., Loraine Hale, John K. Carlson, Enric Cortés, Charles ctenophore Mnemiopsis leidyi. Marine Ecology Progress Series
A. Manire, and James Gelsleichter. 2007. Geographic and ontoge- 280: 163–172. https://doi.org/10.3354/meps280163.
netic variation in the diet and daily ration of the bonnethead shark, Diaz, Robert J., and Rutger Rosenberg. 2008. Spreading dead zones and
Sphyrna tiburo, from the eastern Gulf of Mexico. Marine Biology consequences for marine ecosystems. Science 321 (5891): 926–929.
152 (5): 1009–1020. https://doi.org/10.1007/s00227-007-0728-7. https://doi.org/10.1126/science.1156401.
Bianchi, Thomas S., Steven F. DiMarco, James H. Cowan, Robert D. DiCiccio, Thomas J., and Bradley Efron. 1996. Bootstrap cinfidence
Hetland, Piers Chapman, John W. Day, and Mead A. Allison. intervals. Statistical Science 11: 189–212. https://doi.org/10.1007/
2010. The science of hypoxia in the northern Gulf of Mexico: A sl0869-007-9037-x.
review. Science of the Total Environment 408 (7): 1471–1484. Domenici, Paolo, Cristel Lefrançois, and A. Shingles. 2007. Hypoxia and
https://doi.org/10.1016/j.scitotenv.2009.11.047. the antipredator behaviours of fishes. Philosophical Transactions of
Brandt, Stephen B., and Doran M. Mason. 2003. Effect of nutrient load- the Royal Society B: Biological Sciences 362 (1487): 2105–2121.
ing on Atlantic menhaden (Brevoortia tyrannus) growth rate poten- https://doi.org/10.1098/rstb.2007.2103.
tial in the Patuxent River. Estuaries 26 (2): 298–309. https://doi.org/ Eby, Lisa A., and Larry B. Crowder. 2002. Hypoxia-based habitat com-
10.1007/BF02695968. pression in the Neuse River Estuary: context-dependent shifts in
Brandt, Stephen B., Marco Costantini, Sarah Kolesar, Stuart A. Ludsin, behavioral avoidance thresholds. Canadian Journal of Fisheries
Doran M. Mason, Christopher M. Rae, and Hongyan Zhang. 2011. and Aquatic Sciences 59 (6): 952–965. https://doi.org/10.1139/f02-
Does hypoxia reduce habitat quality for Lake Erie walleye (Sander 067.
vitreus)? A bioenergetics perspective. Canadian Journal of Ekau, Werner, Holger Auel, Hans O. Portner, and Denis Gilbert. 2010.
Fisheries and Aquatic Sciences 68 (5): 857–879. https://doi.org/ Impacts of hypoxia on the structure and processes in pelagic com-
10.1139/f2011-018. munities (zooplankton, macro-invertebrates and fish).
Breitburg, Denise L., Timothy Loher, Carol A. Pacey, and Adam Biogeosciences 7 (5): 1669–1699. https://doi.org/10.5194/bg-7-
Gerstein. 1997. Varying effects of low dissolved oxygen on trophic 1669-2010.
interactions in an estuarine food web. Ecological Monographs 67 Elliott, David T., James J. Pierson, and Michael R. Roman. 2013.
(4): 489–507. Predicting the effects of coastal hypoxia on vital rates of the plank-
Breitburg, Denise L., Aaron T. Adamack, Kenneth A. Rose, Sarah E. tonic copepod Acartia tonsa Dana. PLoS One 8 (5): e63987. https://
Kolesar, Beth Decker, Jennifer E. Purcell, Julie E. Keister, and doi.org/10.1371/journal.pone.0063987.
James H. Cowan. 2003. The pattern and influence of low dissolved Fontaine, Clark T., and Richard A. Neal. 1971. Length-weight relations
oxygen in the Patuxent River, a seasonally hypoxic estuary. for three commercially important penaeid shrimp of the Gulf of
Estuaries 26 (2): 280–297. Mexico. Transactions of the American Fisheries Society 100 (3):
Breitburg, Denise L., J. Kevin Craig, Richard S. Fulford, Kenneth A. 584–586.
Rose, Walter R. Boynton, Damian C. Brady, Benjamin J. Ciotti, Glaspie, Cassandra N., Melissa A. Clouse, Stuart A. Ludsin, Doran M.
et al. 2009. Nutrient enrichment and fisheries exploitation: Mason, Michael R. Roman, Craig A. Stow, and Stephen B. Brandt.
Interactive effects on estuarine living resources and their manage- 2018. Effect of hypoxia on diet of Atlantic Bumpers in the Northern
ment. Hydrobiologia 629 (1): 31–47. https://doi.org/10.1007/ Gulf of Mexico. Transactions of the American Fisheries Society 147
s10750-009-9762-4. (4): 740–748. https://doi.org/10.1002/tafs.10063.
Buchheister, Andre, Christopher F. Bonzek, James Gartland, and Robert Hartigan, John A., and Melissa A. Wong. 1979. A k-means clustering
J. Latour. 2013. Patterns and drivers of the demersal fish community algorithm. Applied Statistics 28: 100–108.
of chesapeake bay. Marine Ecology Progress Series 481: 161–180. Hazen, Elliott L., J. Kevin Craig, Caroline P. Good, and Larry B.
https://doi.org/10.3354/meps10253. Crowder. 2009. Vertical distribution of fish biomass in hypoxic wa-
Cadman, Linda R., and Michael P. Weinstein. 1985. Size-weight relation- ters on the Gulf of Mexico shelf. Marine Ecology Progress Series
ships of postecdysial juvenile blue crabs (Callinectes sapidus 375: 195–207. https://doi.org/10.3354/meps07791.
Rathbun) from the Lower Chesapeake Bay. Journal of Crustacean Hennig, Christian. 2007. Cluster-wise assessment of cluster stability.
Biology 5 (2): 306–310. Computational Statistics and Data Analysis 52 (1): 258–271.
Chisholm, Laurie A., and John C. Roff. 1990. Size-weight relationships https://doi.org/10.1016/j.csda.2006.11.025.
and biomass of tropical neritic copepods off Kingston, Jamaica. Hennig, Christian. 2008. Dissolution point and isolation robustness:
Marine Biology 106 (1): 71–77. Robustness criteria for general cluster analysis methods. Journal of
Costantini, Marco, Stuart A. Ludsin, Doran M. Mason, Xinsheng Zhang, Multivariate Analysis 99 (6): 1154–1176. https://doi.org/10.1016/j.
William C. Boicourt, and Stephen B. Brandt. 2008. Effect of hyp- jmva.2007.07.002.
oxia on habitat quality of striped bass (Morone saxatilis) in Hoffmayer, Eric R., and Glenn R. Parsons. 2003. Food habits of three
Chesapeake Bay. Canadian Journal of Fisheries and Aquatic shark species from the Mississippi Sound in the Northern Gulf of
Sciences 65 (5): 989–1002. https://doi.org/10.1139/f08-021. Mexico. Southeastern Naturalist 2 (2): 271–280.
2182 Estuaries and Coasts (2019) 42:2170–2183

Hopcroft, Russell R., John C. Roff, and Heather A. Bouman. 1998. Pihl, Leif, Susanne P. Baden, Robert J. Diaz, and Linda C. Schaffner.
Zooplankton growth rate: the larvaceans Appendicularia, Fritillaria 1992. Hypoxia-induced structural changes in the diet of bottom-
and Oikopleura in tropical waters. Journal of Plankton Research 20 feeding fish and Crustacea. Marine Biology 112 (3): 349–361.
(3): 539–555. https://doi.org/10.1007/BF00356279.
Hughes, Brent B., Matthew D. Levey, Monique C. Fountain, Aaron B. Pothoven, Steven A., Henry A. Vanderploeg, Thomas O. Höök, and
Carlisle, Francisco P. Chavez, and Mary G. Gleason. 2015. Climate Stephen B. Brandt. 2009. Feeding ecology of emerald shiners and
mediates hypoxic stress on fish diversity and nursery function at the rainbow smelt in central Lake Erie. Journal of Great Lakes
land–sea interface. Proceedings of the National Academy of Resources 35 (2): 190–198.
Sciences 112 (26): 8025–8030. https://doi.org/10.1073/pnas. Prince, Eric D., and C. Phillip Goodyear. 2006. Hypoxia-based habitat
1505815112. compression of tropical pelagic fishes. Fisheries Oceanography 15
Karnauskas, Mandy, Christopher R. Kelble, Seann Regan, Charline (6): 451–464. https://doi.org/10.1111/j.1365-2419.2005.00393.x.
Quenée, Rebecca Allee, Michael Jepson, Amy Freitag, et al. 2017. R Core Team. 2019. R: A language and environment for statistical
2017 Ecosystem Status Report Update for the Gulf of Mexico. computing. Vienna, Austria: R Foundation for Statistical
Keister, Julie E., Edward D. Houde, and Denise L. Breitburg. 2000. Computing.
Effects of bottom-layer hypoxia on abundances and depth distribu- Rabalais, Nancy N., and R. Eugene Turner. 2001. Hypoxia in the northern
tions of organisms in Patuxent River, Chesapeake Bay. Marine Gulf of Mexico: Description, causes and change. In Coastal
Ecology Progress Series 205: 43–59. https://doi.org/10.3354/ Hypoxia: Consequences for Living Resources and Ecosystems, ed.
meps205043. Nancy N. Rabalais and R. Eugene Turner. Washington D. C.:
Kimmel, David G., William C. Boicourt, James J. Pierson, Michael R. American Geophysical Union.
Roman, and Xinsheng Zhang. 2009. A comparison of the Rabotyagov, Sergey S., Todd D. Campbell, Michael White, Jeffrey G.
mesozooplankton response to hypoxia in Chesapeake Bay and the Arnold, Atwood Jay, M. Lee Norfleet, Catherine L. Kling, et al.
northern Gulf of Mexico using the biomass size spectrum. Journal 2014. Cost-effective targeting of conservation investments to reduce
of Experimental Marine Biology and Ecology 381. Elsevier B.V.: the northern Gulf of Mexico hypoxic zone. Proceedings of the
S65–S73. https://doi.org/10.1016/j.jembe.2009.07.012. National Academy of Sciences 111 (52): 18530–18535. https://doi.
Kimmel, David G., William C. Boicourt, James J. Pierson, Michael R. org/10.1073/pnas.1405837111.
Roman, and Xinsheng Zhang. 2010. The vertical distribution and Rahel, Frank J., and Julie W. Nutzman. 1994. Foraging in a lethal envi-
diel variability of mesozooplankton biomass, abundance and size in ronment: Fish predation in hypoxic waters of a stratified lake.
response to hypoxia in the northern Gulf of Mexico USA. Journal of Ecology 75 (5): 1246–1253.
Plankton Research 32 (8): 1185–1202. Remsen, Andrew, Thomas L. Hopkins, and Scott Samson. 2004. What
Langseth, Brian J., Kevin M. Purcell, J. Kevin Craig, Amy M. Schueller, you see is not what you catch: A comparison of concurrently col-
Joseph W. Smith, Kyle W. Shertzer, Sean Creekmore, Kenneth A. lected net, Optical Plankton Counter, and Shadowed Image Particle
Rose, and Katja Fennel. 2014. Effect of changes in dissolved oxygen Profiling Evaluation Recorder data from the northeast Gulf of
concentrations on the spatial dynamics of the Gulf menhaden fishery Mexico. Deep Sea Research Part I: Oceanographic Research
in the Northern Gulf of Mexico. Marine and Coastal Fisheries 6 (1): Papers 51 (1): 129–151.
223–234. https://doi.org/10.1080/19425120.2014.949017. Renaud, Maurice L. 1986. Hypoxia in Louisiana coastal waters during
Long, W. Christopher, and Rochelle D. Seitz. 2008. Trophic interactions 1983: Implications for fisheries. Fishery Bulletin 84 (10): 19–26.
under stress: Hypoxia enhances foraging in an estuarine food web. https://doi.org/10.1063/1.1453944.
Marine Ecology Progress Series 362: 59–68. https://doi.org/10. Roberts, James J., Tomas O. Höök, Stuart A. Ludsin, Steven A.
3354/meps07395. Pothoven, Henry A. Vanderploeg, and Stephen B. Brandt. 2009.
Long, W. Christopher, Bryce J. Brylawski, and Rochelle D. Seitz. 2008. Effects of hypolimnetic hypoxia on foraging and distributions of
Behavioral effects of low dissolved oxygen on the bivalve Macoma Lake Erie yellow perch. Journal of Experimental Marine Biology
balthica. Journal of Experimental Marine Biology and Ecology 359 and Ecology 381: S132–S142. https://doi.org/10.1016/j.jembe.
(1): 34–39. https://doi.org/10.1016/j.jembe.2008.02.013. 2009.07.017.
Ludsin, Stuart A., Xinsheng Zhang, Stephen B. Brandt, Michael R. Roman, Michael R., James J. Pierson, David G. Kimmel, William C.
Roman, William C. Boicourt, Doran M. Mason, and Marco Boicourt, and Xinsheng Zhang. 2012. Impacts of hypoxia on zoo-
Costantini. 2009. Hypoxia-avoidance by planktivorous fish in plankton spatial distributions in the Northern Gulf of Mexico.
Chesapeake Bay: Implications for food web interactions and fish Estuaries and Coasts 35 (5): 1261–1269. https://doi.org/10.1007/
recruitment. Journal of Experimental Marine Biology and Ecology s12237-012-9531-x.
381: 121–131. https://doi.org/10.1016/j.jembe.2009.07.016. Roman, Michael R., Stephen B. Brandt, Edward D. Houde, and James J.
Manooch, C.S., and W.T. Hogarth. 1983. Stomach contents and giant Pierson. 2019. Interactive effects of hypoxia and temperature on
trematodes from wahoo, Acanthocybium solanderi, collected along coastal pelagic zooplankton and fish. Frontiers in Marine Science
the South Atlantic and Gulf coasts of the United States. Bulletin of 6: 139. https://doi.org/10.3389/fmars.2019.00139.
Marine Science 33: 227–238. Rose, Kenneth A., John C. Roff, and Russell R. Hopcroft. 2004.
Meyer, Gabriele H., and James S. Franks. 1996. Food of cobia, Production of Penilia avirostris in Kingston Harbour, Jamaica.
Rachycentron canadum, from the Northcentral Gulf of Mexico. Journal of Plankton Research 26 (6): 605–615.
Gulf Research Reports 9: 161–167. https://doi.org/10.18785/grr. Rose, Kenneth A., Sean B. Creekmore, Dubravko Justić, Thomas Peter, J.
0903.02. Kevin Craig, Rachel Miller Neilan, Lixia Wang, Md S. Rahman, and
NMFS. 2017a. Fisheries Economics of the United States 2015. NOAA David Kidwell. 2018. Modeling the population effects of hypoxia
Technical Memorandum: 247. https://doi.org/10.1017/ on Atlantic croaker (Micropogonias undulatus) in the Northwestern
CBO9781107415324.004. Gulf of Mexico. Estuaries and Coasts 41 (1): 233–254. https://doi.
NMFS. 2017b. Fisheries Economics of the United States 2015. org/10.1007/s12237-017-0266-6.
Washington, D.C.: NOAA Technical Memorandum. https://doi. Shang, Eva H.H., and Rudolf S.S. Wu. 2004. Aquatic hypoxia is a terato-
org/10.1017/CBO9781107415324.004. gen and affects fish embryonic development. Environmental
NMFS. 2018. Fisheries of the United States. https://doi.org/10.2307/j. Science and Technology 38 (18): 4763–4767. https://doi.org/10.
ctv157bc5.9. 1021/es0496423.
Estuaries and Coasts (2019) 42:2170–2183 2183

Stalder, L.C., and Nancy H. Marcus. 1997. Zooplankton responses to Vanderploeg, Henry A., Stuart A. Ludsin, Steven A. Ruberg, Tomas O.
hypoxia: Behavioral patterns and survival of three species of Höök, Steven A. Pothoven, Stephen B. Brandt, Gregory A. Lang,
calanoid copepods. Marine Biology 127 (4): 599–607. James R. Liebig, and Joann F. Cavaletto. 2009b. Hypoxia affects
Sutton, Tracy T., and Thomas L. Hopkins. 1996. Trophic ecology of the spatial distributions and overlap of pelagic fish, zooplankton, and
stomiid (Pisces: Stomiidae) fish assemblage of the eastern Gulf of phytoplankton in Lake Erie. Journal of Experimental Marine
Mexico: Strategies, selectivity and impact of a top mesopelagic Biology and Ecology 381: S92–S107. https://doi.org/10.1016/j.
predator group. Marine Biology 127 (2): 179–192. https://doi.org/ jembe.2009.07.027.
10.1007/BF00942102. Vaquer-Sunyer, Raquel, and Carlos M. Duarte. 2008. Thresholds of hyp-
Taylor, J. Christopher, and Peter S. Rand. 2003. Spatial overlap and oxia for marine biodiversity. Proceedings of the National Academy
distribution of anchovies (Anchoa spp.) and copepods in a shallow of Sciences 105 (1): 15452–15457. https://doi.org/10.4319/lo.2007.
stratified estuary. Aquatic Living Resources 16 (3): 191–196. https:// 52.1.0338.
doi.org/10.1016/S0990-7440(03)00012-3. Webber, Mona K., and John C. Roff. 1995. Annual biomass and produc-
Taylor, J. Christopher, Peter S. Rand, and Jacqueline Jenkins. 2007. tion of the oceanic copepod community off Discovery Bay, Jamaica.
Swimming behavior of juvenile anchovies (Anchoa spp.) in an ep- Marine Biology 123 (3): 481–495.
isodically hypoxic estuary: Implications for individual energetics Wells, R., J. David, James H. Cowan, and Brian Fry. 2008. Feeding
and trophic dynamics. Marine Biology 152 (4): 939–957. ecology of red snapper Lutjanus campechanus in the northern
Thomas, Peter, and Md. Saydur Rahman. 2012. Extensive reproductive Gulf of Mexico. Marine Ecology Progress Series 361: 213–225.
disruption, ovarian masculinization and aromatase suppression in https://doi.org/10.3354/meps07425.
Atlantic croaker in the northern Gulf of Mexico hypoxic zone. Zhang, G.T., and C.K. Wong. 2011. Changes in the planktonic copepod
Proceedings of the Royal Society B: Biological Sciences 279 community in a landlocked bay in the subtropical coastal waters of
(1726): 28–38. https://doi.org/10.1098/rspb.2011.0529. Hong Kong during recovery from eutrophication. Hydrobiologia
Thronson, Amanda, and Antonietta Quigg. 2008. Fifty-five years of fish 666 (1): 277–288. https://doi.org/10.1007/s10750-010-0422-5.
kills in Coastal Texas. Estuaries and Coasts 31 (4): 802–813. https://
Zhang, Hongyan, Stuart A. Ludsin, Doran M. Mason, Aaron T.
doi.org/10.1007/sl2237-008-9056-5.
Adamack, Stephen B. Brandt, Xinsheng Zhang, David G.
Tita, Guglielmo, Magda Vincx, and Gaston Desrosiers. 1999. Size spec-
Kimmel, Michael R. Roman, and William C. Boicourt. 2009.
tra, body width and morphotypes of intertidal nematodes: An eco-
Hypoxia-driven changes in the behavior and spatial distribution of
logical interpretation. Journal of the Marine Biological Association
pelagic fish and mesozooplankton in the northern Gulf of Mexico.
of the United Kingdom 79 (6): 1007–1015.
Journal of Experimental Marine Biology and Ecology 381: S80–
Turner, R. Eugene, Nancy N. Rabalais, and Dubravko Justic. 2008. Gulf
S91. https://doi.org/10.1016/j.jembe.2009.07.014.
of Mexico hypoxia: Alternate states and a legacy. Environmental
Science and Technology 42 (7): 2323–2327. Zhang, Hongyan, Doran M. Mason, Craig A. Stow, Aaron T. Adamack,
Uye, Shin-Ichi. 1982. Length-weight relationships of important zoo- Stephen B. Brandt, Xinsheng Zhang, David G. Kimmel, Michael R.
plankton from the Inland Sea of Japan. Journal of the Roman, William C. Boicourt, and Stuart A. Ludsin. 2014. Effects of
Oceanographical Society of Japan 38 (3): 149–158. hypoxia on habitat quality of pelagic planktivorous fishes in the
Vanderploeg, Henry A., Stuart A. Ludsin, Joann F. Cavaletto, Tomas O. northern gulf of Mexico. Marine Ecology Progress Series 505:
Höök, Steven A. Pothoven, Stephen B. Brandt, James R. Liebig, and 209–226. https://doi.org/10.3354/meps10768.
Gregory A. Lang. 2009a. Hypoxic zones as habitat for zooplankton Zielinski, S., P.G. Lee, and H.O. Portner. 2000. Metabolic performance of
in Lake Erie: Refuges from predation or exclusion zones? Journal of the squid Lolliguncula brevis during hypoxia: An analysis of the
Experimental Marine Biology and Ecology 381: S108–S120. critical PO2. Journal of Experimental Marine Biology and Ecology
https://doi.org/10.1016/j.jembe.2009.07.015. 243 (2): 241–259.